SLR - September 2014 - Michael L. Sganga
Reference: B. Kessler, M. Knupp, P. Graber, L. Zwicky, B. Hintermann, W. Zimmerli, P. Sendi. The Treatment and Outcome of Peri-prosthetic Infection of the Ankle: A Single Cohort-Centre Experience of 34 Cases. Bone Joint Journal June 2014 vol. 96-B. no. 6; 772-777.Scientific Literature Review
Reviewed By: Michael L. Sganga, DPM
Residency Program: Mount Auburn Hospital, Cambridge, MA
Podiatric Relevance: Over recent years, total ankle replacements (TAR) have become a viable option to fusion in the treatment of ankle arthritis. With the emergence of third-generation prostheses and better poly technology, ankle replacements are becoming increasingly common, and along with them come the potential risk of complications. According to the recent IDSA publication, prosthetic joint infections (PJI) are considered a serious complication and this is something that one must understand and manage properly if it were to occur. Published PJIs for TAR range from 2-8.6 percent and this currently exceeds other joint arthroplasties. This article sought to evaluate and compare treatment concepts from a previously published algorithm for hip and knee infected prostheses and PJI involving the ankle.
Methods: A total of 34 patients with PJI diagnosis were retrospectively identified on computer database from a cohort of 511 consecutive patients that had a TAR between August 2006 and June 2011. Of those 24 had a primary TAR. Diagnosis of PJI included clinical exam findings and at least one of the following: 1) growth of the same organism in at least two synovial cultures or peri-prosthetic tissue, 2) visible pus surrounding implant, 3) acute inflammation on histopathologic exam, or 4) direct sinus communicating with TAR. There were four treatments that had been rendered at surgeon discretion: 1) surgery with complete retention or only partial replacement of the components (21 or 61.8 percent), 2) a one or two-stage revision (10 or 29.4 percent), 3) conversion to arthrodesis (3 or 8.8 percent), or 4) amputation (none). This was performed at surgeon discretion and antimicrobial protocols were followed. Endpoints were assessed as: survival without relapse, infection-free survival, and satisfactory function. Analysis also included whether or not the treatment was performed according to the published algorithm. One patient was excluded from this analysis due to having a year’s long suppressive antibiosis and talar component exchange.
The PJI criteria for surgery with retention of one or both components published on the hip and knee are: acute infection, only slightly compromised soft tissue, stable components, and a causative pathogen that is susceptible to an agent with activity against biofilms.
Results: Surgery with retention of one or both components was the most common and resolved infection in 17 or 80.9 percent of the cases. Four of the 21 total went on to residual infections. Of those, three had arthrodesis and one had an amputation; none required amputation as primary procedure. Replacement of all components (10 total) and conversion to arthrodesis (three total) had 100 percent resolution. The conversion to arthrodesis resulted in loss of function. Of the ten that were treated with total revision one had a 1-stage and nine had a 2-stage procedure. The investigators note that the comorbid conditions, type of infection, and duration of symptoms were not associated with the outcome of treatment. They most importantly note that if the treatment strategy deviated from that of the hip or knee the outcomes were not inferior.
Of the 21 PJI’s treated with retention of implant only 4 (19 percent) met the criteria for the published algorithm and they all resolved infection. The remaining 16 (one excluded for long term antibiotics) patients, were treated with retention of one or both components, but did not fulfill criteria for this procedure according to the published algorithm. They were analyzed with a two-tailed Fisher’s Exact test giving p=0.063 showing that two or more criteria not favoring surgery with retention of one or both components was associated with relapse. This was also mirrored when looking at chronic infection. However, 14/17 (82.4 percent) of the remaining patients had a relapse-free survival and 11 (64.7 percent) had an infection-free survival of more than two years. This was in contrast to previous studies on hip, knee, and elbow replacement.
Conclusions: In this study only four patients met the criteria for surgery with retention of one or all components of the prostheses according to the published algorithm for hip and knee implants. Factors associated with failure were having at least two criteria against retention of one or both components (according to published algorithm), chronic infection, or unstable components. The authors do not recommend disregarding the published algorithm, as there is still risk for recurrent infection. However, the grading of soft tissue compromise is too strict with regards to the hip and knee when applied to the ankle joint. According to the study, this did not predict failure of treatment and the authors call for redefining of that with further studies. They cite the different soft tissue envelope as being thinner at the ankle and thus the bacterial load and degree of inflamed tissue is different.